pull-from-upstream job fails for llvm-test-suite

[tstellar]

What happened? What is the problem?

I'm seeing the packit jobs for llvm-test-suite fail. See for example: https://dashboard.packit.dev/jobs/pull-from-upstream/23219

I'm not sure exactly why this is failing, and I cannot reproduce this locally. I'm wondering if maybe the system running the service is running out of disk space.

What did you expect to happen?

No response

Example URL(s)

No response

Steps to reproduce

1.
2.
3.

What is the impacted category (job)?

Fedora release automation

Workaround

• There is an existing workaround that can be used until this issue is fixed.

Participation

• I am willing to submit a pull request for this issue. (Packit team is happy to help!)

[nforro]

I'm wondering if maybe the system running the service is running out of disk space.

It's a bit more complex. The sandbox pod root filesystem has plenty of space available (tens of gigabytes), but the working directory for actions (where $PACKIT_UPSTREAM_REPO and $PACKIT_DOWNSTREAM_REPO also are) is on a volume that has only 3GiB.
We could (and probably will) increase that size, however I think you can work it around by processing the tarball outside the limited volume, for example by setting the working directory for the pkg_test_suite.sh to a /tmp subdirectory or similar.